1 reply to this topic

[go-fish]

I am reviewing vendor compressor discharge PRV calculations. It is a 6 stage recip compressor in hydrogen service. Each stage has a PRV on discharge. The relief scenarios considered are blocked outlet due to either a blocked strainer on the next stage suction or mechanical failure of next stage discharge cylinder valves. There are no valves in the interconnecting discharge piping. So can’t think of any other blocked outlet scenario. In addition, scenario has been considered of interstage recycle valves going wide open from a higher stage to a lower stage. I have a couple of questions:

1. As the stage 2 thru stage 6 PRV discharges are combined into a sub-header, could there be a scenario where more than one stage PRV is relieving? My understanding is it is not credible as I do not foresee see any common mode failure. Recycle valve controller failure and blocked outlet should not be considered simultaneously as independent events. Each recycle valve is sized for 100% recycle.

2. For 3% inlet line losses criteria, should the line start from discharge pulsation dampener up to the PRV inlet or just the PRV inlet line section? PRV is located on the discharge piping about 20 metres from the compressor.

[Pilesar]

My opinions for what they are worth... probably not worth a lot since I tend to argue with designers of my relief valves. Sometimes I win, usually I don't.

I don't see a common mode failure from your description.

The 3% rule doesn't make sense in these scenarios (probably my opinion only) since the cause is mechanical pressuring of non-condensable. Valve chatter seems so unlikely to me here in any reasonable scenario. But if you must conform to 3% inlet rule, then I think it should be calculated from the source unit the PRV is protecting and include the discharge piping.

These opinions are to contribute to discussion and not authoritative.